Flight recorders are certified to withstand impact shock loads of 3,400g for 6.5 milliseconds, applied as a half-sine-wave signal. That figure is a minimum, not a ceiling — survival above it is possible, which is exactly why dramatic numbers spread so easily.
The claim that a Fairchild cockpit voice recorder absorbed an impact estimated above 6,000g is not confirmed in any accident investigation record reviewed for this article. What the record does support is a precise, testable certification regime — and a real gap between what a standard guarantees and what survival folklore assumes.
Somewhere between an engineering tolerance and a retelling, a number hardened into a fact.
The version that circulates is specific: a Fairchild cockpit voice recorder rode out a crash impact estimated at more than 6,000g — roughly double the 3,400g it was built to take. It reads like proof of a hidden engineering margin nobody advertises.
But no documented investigation record reviewed here ties that figure to that recorder. Not an NTSB factual report, not a manufacturer test result, not a named engineer’s estimate. The 6,000g number has the shape of a finding without the paper trail of one.
That absence is the more useful story, because the question underneath it — how much punishment a flight recorder really survives, and who gets to say so — is answered by standards that are precise, testable, and routinely misread.
What the 3,400g figure actually certifies
Start with what the standard says, because nearly everything else follows from it. A crash-protected recorder must withstand impact shock of 3,400g sustained for 6.5 milliseconds, delivered as a half-sine-wave signal — a smooth single-direction pulse, not a ragged crash. The requirement comes from EUROCAE ED-112A, the minimum operational performance specification governing crash-protected recorder systems.
Two details in that sentence do the heavy lifting. The first is the unit: 3,400g is not a force in the everyday sense, it is 3,400 times the pull of gravity held for less than a hundredth of a second.
The second is the word minimum. A recorder that passes at 3,400g is not a recorder that fails at 3,401g. Certifying bodies test to a floor, and a floor is an obligation, not a promise about the ceiling.
Impact shock is only one axis of the requirement. The crash-survivable memory unit must also take compression loads of 5,000 pounds on each orthogonal and diagonal axis — a test run separately, and one that punishes a different failure mode than a violent jolt does.
| Test parameter | Mid-1960s/Fairchild A-100A regime | Modern crash-protected recorder regime | Source status |
|---|---|---|---|
| Impact shock | About 100g for 11ms (historical figure) | 3,400g for 6.5ms, half-sine shock wave | Modern value in ED-112A; A-100A value unverified |
| Static compression | No verified mid-1960s value located | 5,000 lbf on each orthogonal and diagonal axis | Modern value in ED-112A |
| Source: EUROCAE ED-112A | |||
Why the Fairchild claim doesn’t fit the machine
Here is where the arithmetic gets awkward. The recorder named in the claim — the Fairchild A-100A, a workhorse cockpit voice recorder of the tape era — captures at least 30 minutes of analog audio on a continuous-loop tape, according to NTSB material.
That design places it in the mid-1960s generation of recorders, when the impact standard was roughly 100g for 11 milliseconds. Stack that against today’s requirement and you get a ratio of about 1 to 34 — not 1 to 2.
Which means the “nearly double the standard” framing may be measuring a 1960s tape deck against a 2010s rulebook.
If that is what happened, the comparison was never apples to apples, and the drama came from the mismatch rather than the machine.
Worth noting, too: A-100A units have been recovered from serious accidents. American Airlines Flight 965 is the example the record keeps returning to. But nowhere in that account is a 6,000g measurement — the tape came out of the wreckage apparently undamaged by heat or impact, and it was still recording when power was cut, which is a good outcome, not a record-breaking one.
Superdeals cover departures from North America, Europe and Australia.
Flight deals most people never see
Air Traveler Club monitors thousands of routes for pricing anomalies that traditional search engines miss. With ATC Superdeals members save up to 80% on each flight vs. regular fares.
Too good to be true? Check it out yourself. It's free.
Delivered weekly. Unsubscribe anytime.
What verifying a 6,000g claim would actually require
A number like this can be real. It just has a shorter list of ways to be real than most retellings acknowledge.
It could be a manufacturer qualification test result — Fairchild running an A-100A past its certified threshold on a rig, with paperwork to match. It could be an accident investigator’s impact estimate, reconstructed from wreckage geometry and aircraft dynamics, which carries a wide confidence band and belongs to one accident, not to the model. Or it could be a paraphrase that lost its hedge somewhere between a technical report and a headline.
Each of those changes how the figure should be read. An engineer’s estimate from a single crash is not a specification.
A manufacturer’s test result is not an accident finding. And a media paraphrase is neither.
What would settle it is dull and specific: an NTSB factual report naming the impact estimate, or a manufacturer test record. Neither appears in the material reviewed here. That is not proof the survival never happened — it is proof that, at present, nobody can point to the document.
How engineers actually prove a recorder can survive
The modern test is a controlled act of violence. A pneumatically powered cannon fires a 50-pound cartridge into a compressible barrier at high speed, and the recorder rides the resulting shock pulse.
The engineering behind that setup is less about brute force than about repeatability. A half-sine wave lets labs anywhere in the world deliver the same pulse shape, so a pass in one facility means the same thing as a pass in another. Random impact, by contrast, is nearly impossible to certify against.
This is where the gap between a minimum and a maximum lives. A recorder certified at 3,400g survives the pulse the test specifies — a clean, brief, single-direction load.
A real crash delivers something messier: multiple impacts, deformation, fire, immersion. Sometimes that combination is survivable well past the certified number. Sometimes it destroys a recorder that would have passed the rig with room to spare.
The distinction matters because it cuts both ways. A certification figure tells you what the equipment is guaranteed to handle. It says nothing about what it will refuse to survive.
That asymmetry is precisely why a striking number attached to a familiar manufacturer spreads faster than the document that would confirm it.
What this means for you
Recorders exist entirely for the aftermath, which is why recovery is engineered as carefully as survival. When a magnetic tape comes out of a damaged unit, NTSB investigators recover it using specialized playback equipment, and tape splicing is sometimes needed to get audio off a broken loop.
Units that end up in water carry acoustic underwater locator beacons, which help searchers find them before the ocean does its work. And the casing is orange for an unglamorous reason: visibility at a crash site, so a recovery team spots it in seconds rather than hours.
The practical takeaway is blunt. When a dramatic flight-recorder number reaches you — 6,000g, or any figure well past the certified threshold — the useful question is not whether it sounds impressive. It is whether anyone can name the document it came from.
A certified floor is not a limit, and a striking claim is not a finding. The two get confused constantly, and that confusion is the reason pieces like this one have to exist at all.
Reporting by
Key terms
- Crash-protected recorder
- A crash-protected recorder is a flight recorder built to survive severe crash conditions so its stored data can be read afterward. EUROCAE ED-112A sets the modern benchmark: the crash-survivable memory unit must take 3,400g of impact shock for 6.5 milliseconds and 5,000 pounds of compression on each orthogonal and diagonal axis, as two separate tests. In this article, the recorder’s certified threshold is the yardstick the disputed 6,000g claim is measured against, even though that threshold was never meant to mark an upper bound.
- Half-sine wave
- A half-sine wave is a single-direction impact load that rises and falls in the shape of half a sine curve. EUROCAE ED-112A specifies it for the 3,400g, 6.5-millisecond shock test, and modern rigs generate it by firing a 50-pound cartridge into a compressible barrier. In this article, the pulse’s standardized shape is the reason a 3,400g pass is a controlled result, not a prediction about any particular crash.
- Continuous-loop tape
- A continuous-loop tape is a magnetic tape format that runs in an endless loop rather than between two reels. It was the storage medium for the tape era of cockpit voice recorders, including the Fairchild A-100A, which captures at least 30 minutes of analog audio on one. In this article, that medium is why the Fairchild claim belongs to a different engineering generation than the 3,400g standard it is measured against.
- Acoustic underwater locator beacon
- An acoustic underwater locator beacon is a small device mounted on a flight recorder that emits regular acoustic pulses underwater. It helps search teams find a recorder that has gone down in water before the unit is lost to the ocean. In this article, the beacon sits on the recovery side of the story: it does nothing to raise a recorder’s certified limits, only the odds that investigators find it.
Questions? Answers.
How do flight recorders survive a crash?
They are certified against a set of severe conditions rather than one. A crash-protected recorder must take 3,400g of impact shock for 6.5 milliseconds as a half-sine wave, and separately withstand 5,000 pounds of compression on each orthogonal and diagonal axis. Modern testing fires a 50-pound cartridge into a compressible barrier to generate that shock, and the standardized orange casing exists to speed recovery at a crash site.
Are today’s flight recorders tested only for impact shock?
No. Impact shock and compression are distinct tests, run separately because they stress the unit in different ways. A crash-survivable memory unit has to absorb 3,400g for 6.5 milliseconds and also hold up against 5,000 pounds of compression applied on each orthogonal and diagonal axis.
How long can a Fairchild A-100A cockpit voice recorder run, and on what medium does it store audio?
The Fairchild A-100A records at least 30 minutes of analog audio on a continuous-loop tape. That tape-based design places it in the older generation of recorder equipment, well before solid-state memory became standard.
What happens to a recorder’s magnetic tape once it is pulled from a crash site?
NTSB investigators typically recover it and rely on specialized playback equipment to read it, and splicing is sometimes required when the tape has been damaged. If the recorder ended up in water, acoustic underwater locator beacons on the unit help searchers find it in the first place.
